On Stage Design Methods

One curious side of the indie uprising of the last few years is how it changed many of the predetermined truths about game development; such as, the ideal environment for game development being a large team, all subject to the will of an auteur game designer. I was quite enamored with this idea when I was younger, as every egomaniac would, but my outlook changed with the time, more due to the fact that most videogame auters act more as brands than true creative dynamos (with some exceptions, of course), but also because being distant with the nuts-and-bolts development make you lose what actually the medium is all about.

Now, most indie designers who are not of an computing background code their projects more out of practical issues than anything, but I also believe the great majority of them wouldn’t have it any other way; being able to be close and personal with the machinery of the game allow you to become less of a builder of systems, and more of a sculptor of sensation.

A long time ago, I read an article claiming that platformers were the jazz of videogames, due to the presence of “standarts” which could be infinitely improvised upon, leading to new and exciting combinations. While I don’t think that’s entirely through (the 2009-2012 never-ending wave of indie platformers certainly gave everyone some burnout), that’s mostly because of the way they are made.

You see, platformers are usually made with rigid level editors, mostly with some sort of grid, are usually sketched or pre-planned somewhat, and those levels are assembled with some attention to geometrical coherence (the little guy must be higher than those gaps).

Now, this is no criticism; there is a reason why the platformer is such an inexhaustible genre, but it certainly does not make itself too favorable to on-the-spot improvisation. If I were to make a musical comparison, it would be to classical music, due to the interplay between distinct mechanics to be analogous to different instruments, all under a solid direction and an unifying theme (that would be running and jumping).

Then what this has to do with In Extremis and the shmup genre? I would suggest that that genre is closer to the free-flowing, dynamic creation of a jam sessions, due to the freedom to quickly implement ideas as long as it is in accordance to an established timing.

In most shoot’em ups, events happen accordingly to a scripted timeline, rather than spatial triggers or player actions. Because of that, good design in those games are all about establishing a good rhythm and making the most out of it individual sections.

In INX, the method I used to design stages was firstly to decide the songs for each stage (each stage having music of a different genre), then analyzing the moments in which different instruments come in and out, the central melodic motifs, as well as peaks and valleys of intensity in each song. Those moments identified, a loose timeline was built for each stage, signalizing at which times new challenges would be introduced.

And that was it. Most enemy behavior, wing formations and bullet patterns were defined right in the moment, not pre-defined, with constant iteration and experimentation to get the perfect timing. Only the larger mechanics and game modifiers were pre-made, and even then they were greatly altered when stages were completed.

While this method helped to constantly come up with fresh ideas and a sense of continuity for each stage, it was also very work intensive; there is very little enemy reusing in the game, with each stage having a distinct set of exclusive enemies, and since Game Maker does not have any sort of visual timeline control (there is a resource called Timelines, but they are a useless, clunky mess), the stage structuring and subsequent adjustments were made in pure code. The problem is, the initial structure I made was pretty poor, due to my lack of deeper understanding of game programming at the time, which made timing corrections further down the line completely exhausting.


But my struggles with code aside, there is something very musical about a genre in which, once you get the rhythm, you can improvise in any way you want. No wonder one of the most accomplished shmup designers in the world, ZUN of the Touhou series, is also a very talented musician, whose melodies are arguably more popular than the games they soundtrack.

In the end, the lack of a pre-existing plan or guidelines helped to get the right groove for each stage, and I feel a lot of the ideas that blossomed during the project would have no chance within a stricter game plan (I’m not a huge fan of GDDs and extensive documentation). Problem is, eventually you need to cut down that torrent of ideas and decide what will and what will not go in the project, with the risk of it becoming a inconsistent hodge-podge of half-made ideas.

(Which is why the game is still in development)

In a future post, we will see the exact benefits of this method in one of the more interesting parts of shump development: bullet patterns. See y’all there, cosmic warriors!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: